Key Takeaways
- A virtual makeover SDK is a licensed toolkit that lets beauty brands embed virtual makeup, hair, and skin try-on inside their own site or app, instead of sending shoppers off to somebody else's beauty platform.
- Only five of the thirteen makeup SDKs here publish a price at all. Banuba TINT, GlamAR, Perfect Corp's Shopify app, DeepAR, and TWIWT show numbers. Everyone else wants a call first.
- Two names on most 2024-era shortlists have moved. Artifutech has left the beauty industry entirely, and Auglio now trades as TWIWT.
- Nothing here was installed or tested. This is a comparison of published pricing and documented features, with a link next to every claim.
Somebody has told you to add virtual try-on to the product pages, and now you need a vendor. You have maybe a week to come back with a shortlist and a budget number. So you start searching, and every result is a vendor blog that ranks itself first.
That is the real problem with this category. The good comparison pages are written by the companies being compared.
This one is written by a company that does not sell a makeup SDK, which at least removes one thumb from the scale. Rooop makes AI-powered virtual try-on for apparel and accessories. No makeup, no hair, no skin analysis. We are not on this list and we cannot be. What follows is what each vendor publishes on its own pages, read on August 5, 2026.
How we compared these virtual makeover SDKs, and what we did not do
We did not install anything. No trials, no benchmarks, no side-by-side makeup rendering tests, and no scores out of ten. Any article claiming it stress-tested thirteen beauty AR SDKs in a week is telling you a story.
What we did instead: read every vendor's own live pages and record two things. First, does the vendor publish a price, and what does that price actually buy. Second, which beauty categories does it name, in its own words.
Modest, I know. It turns out to be the most useful filter available, because roughly two thirds of this market publishes nothing at all, and the gap between "starts at $49" and "book a demo" is the single biggest predictor of how your next six weeks will go.
We also excluded ratings. There are star scores floating around for some of these virtual try-on tools, and at least one of them rests on a review count you could fit in a taxi. Quoting it would mislead more than it informs.
The order below runs from most published detail to least. It is not a quality ranking, because we have no basis for one, and honestly, neither does anyone else who has not written a check.
What is a virtual makeover SDK?
A virtual makeover SDK is a software development kit that lets beauty brands embed AI and AR try-on for makeup, hair color, or skin analysis directly into their own website, mobile app, or in-store screen. It bundles the face tracking, the makeup rendering, and the product colors into something your developers install, so you do not build computer vision from scratch.
Most kits ship three things. A tracking engine that follows the face. A rendering layer that paints lipstick or hair color onto it convincingly. And a digitization pipeline that turns your physical SKUs into try-on-ready color data.
That third part is the one brands underestimate. A tube of lipstick is not a hex code. Finish, opacity, and how the pigment behaves under a phone camera all have to be captured, and most vendors do that work for you during onboarding.
AR makeup try-on, AI photo try-on, and the difference that matters
Augmented reality makeup try-on runs live. The shopper opens the camera, grants permission, and watches the makeup look move with their face in real time. It is the same class of face tracking that powers face filter apps, tuned for color accuracy rather than entertainment. Most AR technologies in this list work exactly that way.
AI photo try-on works from a still image the shopper uploads. No camera permission, no live tracking, and it works on desktop, where AR often falls apart.
For makeup specifically, live AR usually wins. Shoppers want to turn their head and check the shade in their own bathroom lighting. For apparel and accessories the calculation flips, which is why Rooop went the photo route and stayed out of beauty entirely.
SDK, API, or no-code widget: which one does your team actually need?
An SDK is a library your developers compile into your app. An API is a call your backend makes. A no-code widget is a script tag your ecommerce manager pastes in.
Most beauty brands need the widget and think they need the SDK. If you sell on Shopify and want virtual makeup try-on on product pages, a plugin gets you live in an afternoon. You want a real software development kit when you have your own mobile app, when try-on has to run offline on an in-store mirror, or when you plan to develop a virtual makeup app of your own and app development is already on the roadmap.
The 13 best virtual makeover SDKs and AR makeup tools for beauty brands
1. Banuba TINT, the clearest published pricing in beauty AR
Banuba publishes more of its pricing than anyone else in beauty AR, across three separate routes. The no-code product runs $49 per month for 1,000 try-ons, $99 for 3,500, and $349 for 15,000. The Shopify app is $319, $999, and $1,599 per month for 10,000, 40,000, and 70,000 try-ons. Custom SDK integration is quoted individually.
The Shopify tiers cover makeup, eyewear, contact lenses, and hair dye. Every plan carries a 14-day free trial, and digitization is included in the license rather than billed per SKU.
Worth noticing how steep the jump is from the no-code product to the Shopify app. If you are early, that $49 tier is the cheapest honest way to put AR makeup on a live storefront.
2. GlamAR, published AR try-on tiers with view and SKU caps
GlamAR publishes flat monthly tiers with hard caps on both views and SKUs. Starter is $250 per month for 10,000 monthly views and 50 SKUs, Growth is $350 for 50,000 views and 100 SKUs, and Scale is $450 for 100,000 views and 500 SKUs. AI skin analysis is quoted separately.
Coverage goes beyond makeup: hair and nails, plus eyewear, jewelry, and even furniture. Integration comes as a no-code embed, a Shopify app, an SDK, or an API.
The SKU cap is the thing to check. Fifty SKUs disappears fast once you count every shade in one foundation line.
3. Perfect Corp (YouCam), the widest beauty AR feature set
Perfect Corp is the largest beauty AR vendor here, and its YouCam Shopify app is the one piece it prices in the open. Essential is $379 per month, Premium is $569 per month, both with a 14-day trial, covering up to ten product categories across makeup and eyewear.
The enterprise side is a different animal. A broad AI beauty portfolio spanning makeup, hair, skin and hair diagnostics, and fashion accessories, delivered through REST APIs and SDKs, with pricing by quote. Perfect Corp says it works with more than 800 brand partners, which makes it the default enterprise beauty AR SDK almost by weight.
If your requirement list includes skin diagnostics alongside makeup and skincare, this is the shortest path to covering both from one vendor.
4. DeepAR, the cheapest way to ship an AR makeup prototype
DeepAR is the cheapest way to get an AR makeup prototype in front of stakeholders, and it prices by monthly active users. Free up to 10 MAU with a watermark, $25 per month to 1,000, $100 to 5,000, $500 to 30,000, $1,000 to 100,000, and custom above that.
Read the fine print before you plan a launch on it. DeepAR's beauty component is in free beta, watermarked, and shuts off automatically after three minutes of use. The AR SDK itself covers iOS, Android, HTML5, and macOS.
Great for a prototype or a short campaign. Not where a foundation shade finder should live.
5. TWIWT, formerly Auglio, for small cosmetic catalogs
If you knew this vendor as Auglio, it is now TWIWT. The old domain redirects and the Shopify listing has been renamed. That app has a free tier for 100 unique users per month, then $59, $159, and $369 per month, with product integration billed at $4 per SKU on paid plans.
Coverage is decorative cosmetics: lipstick, foundation, blush, eyeshadow, eyeliner, and highlighter. Shopify and WooCommerce plugins are pre-built, with onboarding support for custom storefronts.
That per-SKU fee is unusual and it is the number to model first. A 300-shade catalog costs $1,200 to digitize before you pay for a single month of service.
6. ModiFace, the L'Oréal-owned incumbent
ModiFace is the incumbent of this whole category, and it is owned by L'Oréal. The technology covers virtual makeup try-on with eleven finishes, hair color, and AI skin diagnostics, and the company reports more than a billion virtual try-on sessions. No pricing is published anywhere.
That ownership is the strategic question, not a technical one. If you compete with a L'Oréal brand, buying try-on from L'Oréal is a conversation for your commercial team rather than your CTO. ModiFace remains the most widely deployed beauty AR technology in the industry, which is exactly why the question matters.
7. Revieve, beauty advice wrapped around the try-on
Revieve wraps beauty advice around the try-on rather than selling the try-on alone. Virtual try-on spans face, lips, eyes, and brow categories plus foundation shade matching, sitting inside a wider recommendation platform. Offices in Helsinki and Chicago, and no published pricing.
Pick this one if the brief is "help brands guide shoppers to the right product", not "let shoppers experiment with makeup".
8. PulpoAR, makeup, nails, and skin in one beauty platform
PulpoAR covers the three things beauty brands usually have to quote separately: makeup, nails, and skin. Live camera, photo upload, and virtual models are all supported, with a smart mirror product for retail floors and an analytics layer on top. Pricing by demo only.
Nail polish try-on is genuinely rare across this list, and it is the reason PulpoAR lands on shortlists it otherwise would not.
9. Orbo AI, makeup try-ons online or in-store
Orbo AI is built for brands that want the same virtual makeup running online or in-store. Lipstick, foundation, and eyeliner try-ons deploy across websites, iOS and Android apps, tablets, digital kiosks, and smart mirrors through an API and SDK, with no separate consumer app required. No published pricing.
The pitch is consistency. One color pipeline, five surfaces, which matters more than it sounds once a shopper sees one shade on a phone and a different one on a counter mirror.
10. mirrAR, beauty plus jewelry, watches, and eyewear
mirrAR has the broadest catalog here, which is either the selling point or the warning sign. Beauty, makeup, and skincare sit alongside jewelry, watches, handbags, and eyewear, delivered through WebAR, a mobile SDK, in-store screens, and social AR filters. Pricing is quote-based.
Good if one vendor has to cover a beauty line and an accessories line for beauty and fashion brands under one roof. Less good if you want a specialist.
11. Visage Technologies, a lightweight on-device makeup SDK
Visage licenses a deliberately lightweight makeup SDK, built to run on modest hardware. Coverage includes foundation, concealer, blush, eyeshadow, mascara, and lipstick, across smartphones, websites, ecommerce platforms, and in-store smart mirrors. A ready-made simulator called Arbelle is offered for brands that want the makeup application without a custom build. Free trial, pricing on request.
This is the developer's pick. It is a component, not a beauty platform, and it behaves like one.
12. BytePlus Effects, mobile-first AR makeup
BytePlus Effects is an AR effects suite with a real makeup module attached. The makeup layer handles lipstick, blusher, contour, pupil coloring, hair dyeing, eyebrows, and eyeshadow, alongside face and body shaping, stickers, and filters, with out-of-the-box iOS and Android SDKs. Free trial available, pricing not published.
Note what is missing. This is mobile-first with no web AR story on the product page. If your traffic is mobile web, ask about that before anything else.
13. Webkul, virtual makeup try-on beyond Shopify
Webkul is the one worth knowing if you are not on Shopify. Its virtual makeup try-on covers lipstick, nail polish, hair color, and bindi, with integrations named for Adobe Commerce, Magento 2, PrestaShop, and Bagisto. No published pricing.
The bindi support is not a rounding error. For brands selling into India, that category is real online sales and almost nobody else renders it.
Virtual makeup SDK pricing compared: who publishes and who does not
| Vendor | Published price | Basis |
|---|---|---|
| Banuba TINT | $49 to $349 no-code; $319 to $1,599 Shopify | try-ons per month |
| GlamAR | $250 to $450 | views and SKUs per month |
| Perfect Corp | $379 to $569 Shopify app | flat, up to 10 categories |
| DeepAR | Free to $1,000 | monthly active users |
| TWIWT | Free to $369, plus $4 per SKU | unique users per month |
| ModiFace | Not published | quote |
| Revieve | Not published | quote |
| PulpoAR | Not published | quote |
| Orbo AI | Not published | quote |
| mirrAR | Not published | quote |
| Visage Technologies | Not published | quote, free trial |
| BytePlus Effects | Not published | quote, free trial |
| Webkul | Not published | quote |
Five vendors publish, eight do not. And the five that publish are not pricing the same thing: Banuba counts try-ons, GlamAR counts views and SKUs, DeepAR and TWIWT count users. A brand with heavy traffic and a small catalog, and a brand with light traffic and 400 shades, will get opposite answers from the same two vendors.
Model your own numbers before you shortlist. Monthly product-page sessions, expected try-on rate, total shade count. Those three figures turn this table into a real budget instead of a wish.
Virtual makeup SDK feature comparison: makeup, hair, nails, and skin tone
| Vendor | Makeup | Hair | Nails | Skin analysis |
|---|---|---|---|---|
| Perfect Corp | Yes | Yes | Not named | Yes |
| Banuba TINT | Yes | Hair dye | Not named | Not named |
| ModiFace | Yes | Yes | Not named | Yes |
| GlamAR | Yes | Yes | Yes | Yes, priced separately |
| Revieve | Yes | Not named | Not named | Recommendations |
| PulpoAR | Yes | Analysis | Yes | Yes |
| Orbo AI | Yes | Not on this page | Not named | Not named |
| mirrAR | Yes | Not named | Not named | Skincare |
| Visage Technologies | Yes | Not named | Not named | Not named |
| DeepAR | Beta | Segmentation | Not named | No |
| BytePlus Effects | Yes | Hair dyeing | Not named | No |
| TWIWT | Yes | Not named | Not named | No |
| Webkul | Yes | Yes | Yes | Not named |
"Not named" means exactly that. The vendor does not list the capability on the pages we read, which is not the same as the capability being absent. Ask.
Skin tone coverage is the gap none of these tables capture. Every vendor claims accuracy across skin tones, and not one publishes evidence you could audit. The only real test is your own: send your five most-returned foundation shades through a demo and see which ones survive. That single check tells you more about makeup simulations than any feature grid will.
Benefits of virtual try-on technology for online beauty brands
The pitch for AI beauty tools is usually delivered as a list of percentages. Ignore the percentages for a minute and look at what actually changes on the page.
A shopper buying makeup online is guessing. They cannot swatch, they cannot hold the tube against their jaw, and the product photo was shot on somebody with a different undertone. Virtual try-on removes the guess, and that is the whole mechanism behind every conversion rate claim in the category.
Three things follow from it. Product pages get longer sessions because trying shades is genuinely fun. Support tickets asking "which shade am I" drop. And shoppers who share their virtual looks bring friends back to the same page, which is why so many vendors bundle social sharing into the shopping experience.
The honest caveat: these are directional benefits, not guaranteed ones. A poorly matched try-on that makes every shade look identical will do worse than no try-on at all, because it teaches shoppers not to trust your colors.
How beauty brands use virtual makeover SDKs online or in-store
Retailer-side tools made this familiar first. Brands like Sephora put makeup try-on in front of millions of shoppers with Sephora's Virtual Artist, and once shoppers had used it once, they started expecting it everywhere.
Four deployment patterns cover almost everything beauty brands use these kits for:
- Product-page try-on, where a shade selector opens a live camera. This is the most common use of virtual try-ons for makeup and the easiest to launch.
- Shade finder quizzes, where the SDK matches foundation to a photo and hands back two or three recommendations.
- In-store smart mirrors and kiosks, where the same catalog runs on a screen at the counter. Perfect Corp, PulpoAR, Orbo AI, mirrAR, and Visage all name this.
- Campaign filters, where an AR makeup look is published for a launch and retired afterwards. DeepAR and BytePlus fit here better than they fit a permanent storefront.
Pick the pattern before the vendor. Two of these thirteen are excellent at pattern four and mediocre at pattern one, and nothing on their pricing page tells you that.
How to choose the right virtual makeover SDK for your beauty brand
Start with the platform, not the features. If you are on Shopify, three of these have a native app and you can be live this month. If you have your own iOS app, that cuts the list down to the ones shipping mobile SDKs. If everything runs through an in-store mirror, you are talking to Perfect Corp, PulpoAR, Orbo AI, mirrAR, or Visage.
Then count your SKUs, because that is where quotes go sideways. Per-SKU digitization at $4 a shade is fine for a 40-product brand and painful for a full color cosmetics line.
Then ask about mobile web. Beauty traffic skews heavily to phones, and a kit that shines inside a native app can struggle in Safari.
And ask what happens to the try-on data. Face imagery is sensitive, rules differ by market, and "we process on device" and "we upload frames to our servers" are very different answers to give your legal team.
The demo question most beauty brands forget to ask
Every vendor demo uses the vendor's own showcase products, tuned over months. Ask them to load three of your SKUs instead, and pick the awkward ones. A sheer gloss. A deep matte foundation. A hair color two shades lighter than the model's roots.
Then watch what happens when the model puts a hand near her face. Makeup that bleeds onto a hand crossing the frame is the tell that tracking is thinner than the marketing suggests. It costs you one email, and it is the single most informative thing you can do before signing anything.
Does virtual makeup try-on actually pay for itself?
Careful with the numbers you will be quoted here. Most conversion-lift figures in this category trace back to vendor case studies with no methodology attached, and the most widely repeated ones are worse than useless.
What is solidly sourced is retail-wide rather than beauty-specific. NRF and Happy Returns put returns at 19.3% of online sales in 2025, and Optoro estimates a return costs roughly 27% of the purchase price.
Beauty is different, though, and this is where most articles get lazy. Hygiene rules mean plenty of cosmetics cannot be returned at all, so the returns argument that works for apparel is weaker here. The real beauty case is sampling cost and shade confidence. Fewer sachets shipped, fewer wrong-shade purchases that end in a one-star review and a customer who never comes back.
Build your business case on that, and on the conversion rate of product pages with try-on measured against those without, on your own traffic, over a full month. Not on a number from a vendor deck.
Where Rooop fits, and where it does not
Full disclosure, and it cuts against us. Rooop is our product, and it does not belong on this list. Rooop does AI virtual try-on for apparel and accessories, from a photo the shopper uploads. No camera permission, no AR, no app. It does not do makeup, hair color, or skin analysis, and we are not going to pretend otherwise on a page about makeup SDKs.
Where it does overlap: beauty brands often sell more than beauty. Merch, hair accessories, sunglasses, bags. If that is a growing slice of your catalog, Rooop covers it at $24, $119, or $399 per month with 20% off annual billing, at 99% visual fidelity, and typically goes live in two to four hours.
For the makeup itself, pick from the thirteen above.